Closed. This question needs to be more focused. It is not currently accepting answers.
Want to improve this question? Update the question so it focuses on one problem only by editing this post.
Closed 4 years ago.
Improve this question
What is an efficient way to implement a singleton design pattern in Java?
Use an enum:
public enum Foo {
INSTANCE;
}
Joshua Bloch explained this approach in his Effective Java Reloaded talk at Google I/O 2008: link to video. Also see slides 30-32 of his presentation (effective_java_reloaded.pdf):
The Right Way to Implement a Serializable Singleton
public enum Elvis {
INSTANCE;
private final String[] favoriteSongs =
{ "Hound Dog", "Heartbreak Hotel" };
public void printFavorites() {
System.out.println(Arrays.toString(favoriteSongs));
}
}
Edit: An online portion of "Effective Java" says:
"This approach is functionally equivalent to the public field approach, except that it is more concise, provides the serialization machinery for free, and provides an ironclad guarantee against multiple instantiation, even in the face of sophisticated serialization or reflection attacks. While this approach has yet to be widely adopted, a single-element enum type is the best way to implement a singleton."
Depending on the usage, there are several "correct" answers.
Since Java 5, the best way to do it is to use an enum:
public enum Foo {
INSTANCE;
}
Pre Java 5, the most simple case is:
public final class Foo {
private static final Foo INSTANCE = new Foo();
private Foo() {
if (INSTANCE != null) {
throw new IllegalStateException("Already instantiated");
}
}
public static Foo getInstance() {
return INSTANCE;
}
public Object clone() throws CloneNotSupportedException{
throw new CloneNotSupportedException("Cannot clone instance of this class");
}
}
Let's go over the code. First, you want the class to be final. In this case, I've used the final keyword to let the users know it is final. Then you need to make the constructor private to prevent users to create their own Foo. Throwing an exception from the constructor prevents users to use reflection to create a second Foo. Then you create a private static final Foo field to hold the only instance, and a public static Foo getInstance() method to return it. The Java specification makes sure that the constructor is only called when the class is first used.
When you have a very large object or heavy construction code and also have other accessible static methods or fields that might be used before an instance is needed, then and only then you need to use lazy initialization.
You can use a private static class to load the instance. The code would then look like:
public final class Foo {
private static class FooLoader {
private static final Foo INSTANCE = new Foo();
}
private Foo() {
if (FooLoader.INSTANCE != null) {
throw new IllegalStateException("Already instantiated");
}
}
public static Foo getInstance() {
return FooLoader.INSTANCE;
}
}
Since the line private static final Foo INSTANCE = new Foo(); is only executed when the class FooLoader is actually used, this takes care of the lazy instantiation, and is it guaranteed to be thread safe.
When you also want to be able to serialize your object you need to make sure that deserialization won't create a copy.
public final class Foo implements Serializable {
private static final long serialVersionUID = 1L;
private static class FooLoader {
private static final Foo INSTANCE = new Foo();
}
private Foo() {
if (FooLoader.INSTANCE != null) {
throw new IllegalStateException("Already instantiated");
}
}
public static Foo getInstance() {
return FooLoader.INSTANCE;
}
#SuppressWarnings("unused")
private Foo readResolve() {
return FooLoader.INSTANCE;
}
}
The method readResolve() will make sure the only instance will be returned, even when the object was serialized in a previous run of your program.
Disclaimer: I have just summarized all of the awesome answers and wrote it in my own words.
While implementing Singleton we have two options:
Lazy loading
Early loading
Lazy loading adds bit overhead (lots of to be honest), so use it only when you have a very large object or heavy construction code and also have other accessible static methods or fields that might be used before an instance is needed, then and only then you need to use lazy initialization. Otherwise, choosing early loading is a good choice.
The most simple way of implementing a singleton is:
public class Foo {
// It will be our sole hero
private static final Foo INSTANCE = new Foo();
private Foo() {
if (INSTANCE != null) {
// SHOUT
throw new IllegalStateException("Already instantiated");
}
}
public static Foo getInstance() {
return INSTANCE;
}
}
Everything is good except it's an early loaded singleton. Lets try lazy loaded singleton
class Foo {
// Our now_null_but_going_to_be sole hero
private static Foo INSTANCE = null;
private Foo() {
if (INSTANCE != null) {
// SHOUT
throw new IllegalStateException("Already instantiated");
}
}
public static Foo getInstance() {
// Creating only when required.
if (INSTANCE == null) {
INSTANCE = new Foo();
}
return INSTANCE;
}
}
So far so good, but our hero will not survive while fighting alone with multiple evil threads who want many many instance of our hero.
So let’s protect it from evil multi threading:
class Foo {
private static Foo INSTANCE = null;
// TODO Add private shouting constructor
public static Foo getInstance() {
// No more tension of threads
synchronized (Foo.class) {
if (INSTANCE == null) {
INSTANCE = new Foo();
}
}
return INSTANCE;
}
}
But it is not enough to protect out hero, really!!! This is the best we can/should do to help our hero:
class Foo {
// Pay attention to volatile
private static volatile Foo INSTANCE = null;
// TODO Add private shouting constructor
public static Foo getInstance() {
if (INSTANCE == null) { // Check 1
synchronized (Foo.class) {
if (INSTANCE == null) { // Check 2
INSTANCE = new Foo();
}
}
}
return INSTANCE;
}
}
This is called the "double-checked locking idiom". It's easy to forget the volatile statement and difficult to understand why it is necessary.
For details: The "Double-Checked Locking is Broken" Declaration
Now we are sure about evil threads, but what about the cruel serialization? We have to make sure even while de-serialiaztion no new object is created:
class Foo implements Serializable {
private static final long serialVersionUID = 1L;
private static volatile Foo INSTANCE = null;
// The rest of the things are same as above
// No more fear of serialization
#SuppressWarnings("unused")
private Object readResolve() {
return INSTANCE;
}
}
The method readResolve() will make sure the only instance will be returned, even when the object was serialized in a previous run of our program.
Finally, we have added enough protection against threads and serialization, but our code is looking bulky and ugly. Let’s give our hero a makeover:
public final class Foo implements Serializable {
private static final long serialVersionUID = 1L;
// Wrapped in a inner static class so that loaded only when required
private static class FooLoader {
// And no more fear of threads
private static final Foo INSTANCE = new Foo();
}
// TODO add private shouting construcor
public static Foo getInstance() {
return FooLoader.INSTANCE;
}
// Damn you serialization
#SuppressWarnings("unused")
private Foo readResolve() {
return FooLoader.INSTANCE;
}
}
Yes, this is our very same hero :)
Since the line private static final Foo INSTANCE = new Foo(); is only executed when the class FooLoader is actually used, this takes care of the lazy instantiation, and is it guaranteed to be thread-safe.
And we have come so far. Here is the best way to achieve everything we did is best possible way:
public enum Foo {
INSTANCE;
}
Which internally will be treated like
public class Foo {
// It will be our sole hero
private static final Foo INSTANCE = new Foo();
}
That's it! No more fear of serialization, threads and ugly code. Also ENUMS singleton are lazily initialized.
This approach is functionally equivalent to the public field approach,
except that it is more concise, provides the serialization machinery
for free, and provides an ironclad guarantee against multiple
instantiation, even in the face of sophisticated serialization or
reflection attacks. While this approach has yet to be widely adopted,
a single-element enum type is the best way to implement a singleton.
-Joshua Bloch in "Effective Java"
Now you might have realized why ENUMS are considered as best way to implement a singleton and thanks for your patience :)
Updated it on my blog.
The solution posted by Stu Thompson is valid in Java 5.0 and later. But I would prefer not to use it because I think it is error prone.
It's easy to forget the volatile statement and difficult to understand why it is necessary. Without the volatile this code would not be thread safe any more due to the double-checked locking antipattern. See more about this in paragraph 16.2.4 of Java Concurrency in Practice. In short: This pattern (prior to Java 5.0 or without the volatile statement) could return a reference to the Bar object that is (still) in an incorrect state.
This pattern was invented for performance optimization. But this is really not a real concern any more. The following lazy initialization code is fast and - more importantly - easier to read.
class Bar {
private static class BarHolder {
public static Bar bar = new Bar();
}
public static Bar getBar() {
return BarHolder.bar;
}
}
Thread safe in Java 5+:
class Foo {
private static volatile Bar bar = null;
public static Bar getBar() {
if (bar == null) {
synchronized(Foo.class) {
if (bar == null)
bar = new Bar();
}
}
return bar;
}
}
Pay attention to the volatile modifier here. :) It is important because without it, other threads are not guaranteed by the JMM (Java Memory Model) to see changes to its value. The synchronization does not take care of that--it only serializes access to that block of code.
#Bno's answer details the approach recommended by Bill Pugh (FindBugs) and is arguable better. Go read and vote up his answer too.
Forget lazy initialization; it's too problematic. This is the simplest solution:
public class A {
private static final A INSTANCE = new A();
private A() {}
public static A getInstance() {
return INSTANCE;
}
}
Make sure that you really need it. Do a google search for "singleton anti-pattern" to see some arguments against it.
There's nothing inherently wrong with it I suppose, but it's just a mechanism for exposing some global resource/data so make sure that this is the best way. In particular, I've found dependency injection (DI) more useful particularly if you are also using unit tests, because DI allows you to use mocked resources for testing purposes.
I'm mystified by some of the answers that suggest dependency injection (DI) as an alternative to using singletons; these are unrelated concepts. You can use DI to inject either singleton or non-singleton (e.g., per-thread) instances. At least this is true if you use Spring 2.x, I can't speak for other DI frameworks.
So my answer to the OP would be (in all but the most trivial sample code) to:
Use a DI framework like Spring Framework, then
Make it part of your DI configuration whether your dependencies are singletons, request scoped, session scoped, or whatever.
This approach gives you a nice decoupled (and therefore flexible and testable) architecture where whether to use a singleton is an easily reversible implementation detail (provided any singletons you use are threadsafe, of course).
Really consider why you need a singleton before writing it. There is a quasi-religious debate about using them which you can quite easily stumble over if you google singletons in Java.
Personally, I try to avoid singletons as often as possible for many reasons, again most of which can be found by googling singletons. I feel that quite often singletons are abused because they're easy to understand by everybody. They're used as a mechanism for getting "global" data into an OO design and they are used because it is easy to circumvent object lifecycle management (or really thinking about how you can do A from inside B). Look at things like inversion of control (IoC) or dependency injection (DI) for a nice middle ground.
If you really need one then Wikipedia has a good example of a proper implementation of a singleton.
Following are three different approaches
Enum
/**
* Singleton pattern example using Java Enum
*/
public enum EasySingleton {
INSTANCE;
}
Double checked locking / lazy loading
/**
* Singleton pattern example with Double checked Locking
*/
public class DoubleCheckedLockingSingleton {
private static volatile DoubleCheckedLockingSingleton INSTANCE;
private DoubleCheckedLockingSingleton() {}
public static DoubleCheckedLockingSingleton getInstance() {
if(INSTANCE == null) {
synchronized(DoubleCheckedLockingSingleton.class) {
// Double checking Singleton instance
if(INSTANCE == null) {
INSTANCE = new DoubleCheckedLockingSingleton();
}
}
}
return INSTANCE;
}
}
Static factory method
/**
* Singleton pattern example with static factory method
*/
public class Singleton {
// Initialized during class loading
private static final Singleton INSTANCE = new Singleton();
// To prevent creating another instance of 'Singleton'
private Singleton() {}
public static Singleton getSingleton() {
return INSTANCE;
}
}
There is a lot of nuance around implementing a singleton. The holder pattern can not be used in many situations. And IMO when using a volatile - you should also use a local variable. Let's start at the beginning and iterate on the problem. You'll see what I mean.
The first attempt might look something like this:
public class MySingleton {
private static MySingleton INSTANCE;
public static MySingleton getInstance() {
if (INSTANCE == null) {
INSTANCE = new MySingleton();
}
return INSTANCE;
}
...
}
Here we have the MySingleton class which has a private static member called INSTANCE, and a public static method called getInstance(). The first time getInstance() is called, the INSTANCE member is null. The flow will then fall into the creation condition and create a new instance of the MySingleton class. Subsequent calls to getInstance() will find that the INSTANCE variable is already set, and therefore not create another MySingleton instance. This ensures there is only one instance of MySingleton which is shared among all callers of getInstance().
But this implementation has a problem. Multi-threaded applications will have a race condition on the creation of the single instance. If multiple threads of execution hit the getInstance() method at (or around) the same time, they will each see the INSTANCE member as null. This will result in each thread creating a new MySingleton instance and subsequently setting the INSTANCE member.
private static MySingleton INSTANCE;
public static synchronized MySingleton getInstance() {
if (INSTANCE == null) {
INSTANCE = new MySingleton();
}
return INSTANCE;
}
Here we’ve used the synchronized keyword in the method signature to synchronize the getInstance() method. This will certainly fix our race condition. Threads will now block and enter the method one at a time. But it also creates a performance problem. Not only does this implementation synchronize the creation of the single instance; it synchronizes all calls to getInstance(), including reads. Reads do not need to be synchronized as they simply return the value of INSTANCE. Since reads will make up the bulk of our calls (remember, instantiation only happens on the first call), we will incur an unnecessary performance hit by synchronizing the entire method.
private static MySingleton INSTANCE;
public static MySingleton getInstance() {
if (INSTANCE == null) {
synchronize(MySingleton.class) {
INSTANCE = new MySingleton();
}
}
return INSTANCE;
}
Here we’ve moved synchronization from the method signature, to a synchronized block that wraps the creation of the MySingleton instance. But does this solve our problem? Well, we are no longer blocking on reads, but we’ve also taken a step backward. Multiple threads will hit the getInstance() method at or around the same time and they will all see the INSTANCE member as null.
They will then hit the synchronized block where one will obtain the lock and create the instance. When that thread exits the block, the other threads will contend for the lock, and one by one each thread will fall through the block and create a new instance of our class. So we are right back where we started.
private static MySingleton INSTANCE;
public static MySingleton getInstance() {
if (INSTANCE == null) {
synchronized(MySingleton.class) {
if (INSTANCE == null) {
INSTANCE = createInstance();
}
}
}
return INSTANCE;
}
Here we issue another check from inside the block. If the INSTANCE member has already been set, we’ll skip initialization. This is called double-checked locking.
This solves our problem of multiple instantiation. But once again, our solution has presented another challenge. Other threads might not “see” that the INSTANCE member has been updated. This is because of how Java optimizes memory operations.
Threads copy the original values of variables from main memory into the CPU’s cache. Changes to values are then written to, and read from, that cache. This is a feature of Java designed to optimize performance. But this creates a problem for our singleton implementation. A second thread — being processed by a different CPU or core, using a different cache — will not see the changes made by the first. This will cause the second thread to see the INSTANCE member as null forcing a new instance of our singleton to be created.
private static volatile MySingleton INSTANCE;
public static MySingleton getInstance() {
if (INSTANCE == null) {
synchronized(MySingleton.class) {
if (INSTANCE == null) {
INSTANCE = createInstance();
}
}
}
return INSTANCE;
}
We solve this by using the volatile keyword on the declaration of the INSTANCE member. This will tell the compiler to always read from, and write to, main memory, and not the CPU cache.
But this simple change comes at a cost. Because we are bypassing the CPU cache, we will take a performance hit each time we operate on the volatile INSTANCE member — which we do four times. We double-check existence (1 and 2), set the value (3), and then return the value (4). One could argue that this path is the fringe case as we only create the instance during the first call of the method. Perhaps a performance hit on creation is tolerable. But even our main use-case, reads, will operate on the volatile member twice. Once to check existence, and again to return its value.
private static volatile MySingleton INSTANCE;
public static MySingleton getInstance() {
MySingleton result = INSTANCE;
if (result == null) {
synchronized(MySingleton.class) {
result = INSTANCE;
if (result == null) {
INSTANCE = result = createInstance();
}
}
}
return result;
}
Since the performance hit is due to operating directly on the volatile member, let’s set a local variable to the value of the volatile and operate on the local variable instead. This will decrease the number of times we operate on the volatile, thereby reclaiming some of our lost performance. Note that we have to set our local variable again when we enter the synchronized block. This ensures it is up to date with any changes that occurred while we were waiting for the lock.
I wrote an article about this recently. Deconstructing The Singleton. You can find more information on these examples and an example of the "holder" pattern there. There is also a real-world example showcasing the double-checked volatile approach.
I use the Spring Framework to manage my singletons.
It doesn't enforce the "singleton-ness" of the class (which you can't really do anyway if there are multiple class loaders involved), but it provides a really easy way to build and configure different factories for creating different types of objects.
Wikipedia has some examples of singletons, also in Java. The Java 5 implementation looks pretty complete, and is thread-safe (double-checked locking applied).
Version 1:
public class MySingleton {
private static MySingleton instance = null;
private MySingleton() {}
public static synchronized MySingleton getInstance() {
if(instance == null) {
instance = new MySingleton();
}
return instance;
}
}
Lazy loading, thread safe with blocking, low performance because of synchronized.
Version 2:
public class MySingleton {
private MySingleton() {}
private static class MySingletonHolder {
public final static MySingleton instance = new MySingleton();
}
public static MySingleton getInstance() {
return MySingletonHolder.instance;
}
}
Lazy loading, thread safe with non-blocking, high performance.
If you do not need lazy loading then simply try:
public class Singleton {
private final static Singleton INSTANCE = new Singleton();
private Singleton() {}
public static Singleton getInstance() { return Singleton.INSTANCE; }
protected Object clone() {
throw new CloneNotSupportedException();
}
}
If you want lazy loading and you want your singleton to be thread-safe, try the double-checking pattern:
public class Singleton {
private static Singleton instance = null;
private Singleton() {}
public static Singleton getInstance() {
if(null == instance) {
synchronized(Singleton.class) {
if(null == instance) {
instance = new Singleton();
}
}
}
return instance;
}
protected Object clone() {
throw new CloneNotSupportedException();
}
}
As the double checking pattern is not guaranteed to work (due to some issue with compilers, I don't know anything more about that), you could also try to synchronize the whole getInstance-method or create a registry for all your singletons.
I would say an enum singleton.
Singleton using an enum in Java is generally a way to declare an enum singleton. An enum singleton may contain instance variables and instance methods. For simplicity's sake, also note that if you are using any instance method then you need to ensure thread safety of that method if at all it affects the state of object.
The use of an enum is very easy to implement and has no drawbacks regarding serializable objects, which have to be circumvented in the other ways.
/**
* Singleton pattern example using a Java Enum
*/
public enum Singleton {
INSTANCE;
public void execute (String arg) {
// Perform operation here
}
}
You can access it by Singleton.INSTANCE, and it is much easier than calling the getInstance() method on Singleton.
1.12 Serialization of Enum Constants
Enum constants are serialized differently than ordinary serializable or externalizable objects. The serialized form of an enum constant consists solely of its name; field values of the constant are not present in the form. To serialize an enum constant, ObjectOutputStream writes the value returned by the enum constant's name method. To deserialize an enum constant, ObjectInputStream reads the constant name from the stream; the deserialized constant is then obtained by calling the java.lang.Enum.valueOf method, passing the constant's enum type along with the received constant name as arguments. Like other serializable or externalizable objects, enum constants can function as the targets of back references appearing subsequently in the serialization stream.
The process by which enum constants are serialized cannot be customized: any class-specific writeObject, readObject, readObjectNoData, writeReplace, and readResolve methods defined by enum types are ignored during serialization and deserialization. Similarly, any serialPersistentFields or serialVersionUID field declarations are also ignored--all enum types have a fixed serialVersionUID of 0L. Documenting serializable fields and data for enum types is unnecessary, since there is no variation in the type of data sent.
Quoted from Oracle documentation
Another problem with conventional Singletons are that once you implement the Serializable interface, they no longer remain singleton because the readObject() method always return a new instance, like a constructor in Java. This can be avoided by using readResolve() and discarding the newly created instance by replacing with a singleton like below:
// readResolve to prevent another instance of Singleton
private Object readResolve(){
return INSTANCE;
}
This can become even more complex if your singleton class maintains state, as you need to make them transient, but with in an enum singleton, serialization is guaranteed by the JVM.
Good Read
Singleton Pattern
Enums, Singletons and Deserialization
Double-checked locking and the Singleton pattern
There are four ways to create a singleton in Java.
Eager initialization singleton
public class Test {
private static final Test test = new Test();
private Test() {
}
public static Test getTest() {
return test;
}
}
Lazy initialization singleton (thread safe)
public class Test {
private static volatile Test test;
private Test() {
}
public static Test getTest() {
if(test == null) {
synchronized(Test.class) {
if(test == null) {
test = new Test();
}
}
}
return test;
}
}
Bill Pugh singleton with holder pattern (preferably the best one)
public class Test {
private Test() {
}
private static class TestHolder {
private static final Test test = new Test();
}
public static Test getInstance() {
return TestHolder.test;
}
}
Enum singleton
public enum MySingleton {
INSTANCE;
private MySingleton() {
System.out.println("Here");
}
}
This is how to implement a simple singleton:
public class Singleton {
// It must be static and final to prevent later modification
private static final Singleton INSTANCE = new Singleton();
/** The constructor must be private to prevent external instantiation */
private Singleton(){}
/** The public static method allowing to get the instance */
public static Singleton getInstance() {
return INSTANCE;
}
}
This is how to properly lazy create your singleton:
public class Singleton {
// The constructor must be private to prevent external instantiation
private Singleton(){}
/** The public static method allowing to get the instance */
public static Singleton getInstance() {
return SingletonHolder.INSTANCE;
}
/**
* The static inner class responsible for creating your instance only on demand,
* because the static fields of a class are only initialized when the class
* is explicitly called and a class initialization is synchronized such that only
* one thread can perform it, this rule is also applicable to inner static class
* So here INSTANCE will be created only when SingletonHolder.INSTANCE
* will be called
*/
private static class SingletonHolder {
private static final Singleton INSTANCE = new Singleton();
}
}
You need the double-checking idiom if you need to load the instance variable of a class lazily. If you need to load a static variable or a singleton lazily, you need the initialization on demand holder idiom.
In addition, if the singleton needs to be serializable, all other fields need to be transient and readResolve() method needs to be implemented in order to maintain the singleton object invariant. Otherwise, each time the object is deserialized, a new instance of the object will be created. What readResolve() does is replace the new object read by readObject(), which forced that new object to be garbage collected as there is no variable referring to it.
public static final INSTANCE == ....
private Object readResolve() {
return INSTANCE; // Original singleton instance.
}
Various ways to make a singleton object:
As per Joshua Bloch - Enum would be the best.
You can use double check locking also.
Even an inner static class can be used.
Enum singleton
The simplest way to implement a singleton that is thread-safe is using an Enum:
public enum SingletonEnum {
INSTANCE;
public void doSomething(){
System.out.println("This is a singleton");
}
}
This code works since the introduction of Enum in Java 1.5
Double checked locking
If you want to code a “classic” singleton that works in a multithreaded environment (starting from Java 1.5) you should use this one.
public class Singleton {
private static volatile Singleton instance = null;
private Singleton() {
}
public static Singleton getInstance() {
if (instance == null) {
synchronized (Singleton.class){
if (instance == null) {
instance = new Singleton();
}
}
}
return instance;
}
}
This is not thread-safe before 1.5 because the implementation of the volatile keyword was different.
Early loading singleton (works even before Java 1.5)
This implementation instantiates the singleton when the class is loaded and provides thread safety.
public class Singleton {
private static final Singleton instance = new Singleton();
private Singleton() {
}
public static Singleton getInstance() {
return instance;
}
public void doSomething(){
System.out.println("This is a singleton");
}
}
For JSE 5.0 and above, take the Enum approach. Otherwise, use the static singleton holder approach ((a lazy loading approach described by Bill Pugh). The latter solution is also thread-safe without requiring special language constructs (i.e., volatile or synchronized).
Another argument often used against singletons is their testability problems. Singletons are not easily mockable for testing purposes. If this turns out to be a problem, I like to make the following slight modification:
public class SingletonImpl {
private static SingletonImpl instance;
public static SingletonImpl getInstance() {
if (instance == null) {
instance = new SingletonImpl();
}
return instance;
}
public static void setInstance(SingletonImpl impl) {
instance = impl;
}
public void a() {
System.out.println("Default Method");
}
}
The added setInstance method allows setting a mockup implementation of the singleton class during testing:
public class SingletonMock extends SingletonImpl {
#Override
public void a() {
System.out.println("Mock Method");
}
}
This also works with early initialization approaches:
public class SingletonImpl {
private static final SingletonImpl instance = new SingletonImpl();
private static SingletonImpl alt;
public static void setInstance(SingletonImpl inst) {
alt = inst;
}
public static SingletonImpl getInstance() {
if (alt != null) {
return alt;
}
return instance;
}
public void a() {
System.out.println("Default Method");
}
}
public class SingletonMock extends SingletonImpl {
#Override
public void a() {
System.out.println("Mock Method");
}
}
This has the drawback of exposing this functionality to the normal application too. Other developers working on that code could be tempted to use the ´setInstance´ method to alter a specific function and thus changing the whole application behaviour, and therefore this method should contain at least a good warning in its javadoc.
Still, for the possibility of mockup-testing (when needed), this code exposure may be an acceptable price to pay.
Simplest singleton class:
public class Singleton {
private static Singleton singleInstance = new Singleton();
private Singleton() {}
public static Singleton getSingleInstance() {
return singleInstance;
}
}
Have a look at this post.
Examples of GoF Design Patterns in Java's core libraries
From the best answer's "Singleton" section,
Singleton (recognizeable by creational methods returning the same instance (usually of itself) everytime)
java.lang.Runtime#getRuntime()
java.awt.Desktop#getDesktop()
java.lang.System#getSecurityManager()
You can also learn the example of Singleton from Java native classes themselves.
The best singleton pattern I've ever seen uses the Supplier interface.
It's generic and reusable
It supports lazy initialization
It's only synchronized until it has been initialized, then the blocking supplier is replaced with a non-blocking supplier.
See below:
public class Singleton<T> implements Supplier<T> {
private boolean initialized;
private Supplier<T> singletonSupplier;
public Singleton(T singletonValue) {
this.singletonSupplier = () -> singletonValue;
}
public Singleton(Supplier<T> supplier) {
this.singletonSupplier = () -> {
// The initial supplier is temporary; it will be replaced after initialization
synchronized (supplier) {
if (!initialized) {
T singletonValue = supplier.get();
// Now that the singleton value has been initialized,
// replace the blocking supplier with a non-blocking supplier
singletonSupplier = () -> singletonValue;
initialized = true;
}
return singletonSupplier.get();
}
};
}
#Override
public T get() {
return singletonSupplier.get();
}
}
I still think after Java 1.5, enum is the best available singleton implementation available as it also ensures that, even in the multi threaded environments, only one instance is created.
public enum Singleton {
INSTANCE;
}
And you are done!
Sometimes a simple "static Foo foo = new Foo();" is not enough. Just think of some basic data insertion you want to do.
On the other hand you would have to synchronize any method that instantiates the singleton variable as such. Synchronisation is not bad as such, but it can lead to performance issues or locking (in very very rare situations using this example. The solution is
public class Singleton {
private static Singleton instance = null;
static {
instance = new Singleton();
// do some of your instantiation stuff here
}
private Singleton() {
if(instance!=null) {
throw new ErrorYouWant("Singleton double-instantiation, should never happen!");
}
}
public static getSingleton() {
return instance;
}
}
Now what happens? The class is loaded via the class loader. Directly after the class was interpreted from a byte Array, the VM executes the static { } - block. that's the whole secret: The static-block is only called once, the time the given class (name) of the given package is loaded by this one class loader.
public class Singleton {
private static final Singleton INSTANCE = new Singleton();
private Singleton() {
if (INSTANCE != null)
throw new IllegalStateException(“Already instantiated...”);
}
public synchronized static Singleton getInstance() {
return INSTANCE;
}
}
As we have added the Synchronized keyword before getInstance, we have avoided the race condition in the case when two threads call the getInstance at the same time.
I read this question about how to do Double-checked locking:
// Double-check idiom for lazy initialization of instance fields
private volatile FieldType field;
FieldType getField() {
FieldType result = field;
if (result == null) { // First check (no locking)
synchronized(this) {
result = field;
if (result == null) // Second check (with locking)
field = result = computeFieldValue();
}
}
return result;
}
My aim is to get lazy-loading a field (NOT a singleton) work without the volatile attribute. The field object is never changed after initialization.
After some testing my final approach:
private FieldType field;
FieldType getField() {
if (field == null) {
synchronized(this) {
if (field == null)
field = Publisher.publish(computeFieldValue());
}
}
return fieldHolder.field;
}
public class Publisher {
public static <T> T publish(T val){
return new Publish<T>(val).get();
}
private static class Publish<T>{
private final T val;
public Publish(T val) {
this.val = val;
}
public T get(){
return val;
}
}
}
The benefit is possibly faster access time due to not needing volatile, while still keeping the simplicity with the reusable Publisher class.
I tested this using jcstress. SafeDCLFinal worked as expected while UnsafeDCLFinal was inconsistent (as expected). At this point im 99% sure it works, but please, prove me wrong. Compiled with mvn clean install -pl tests-custom -am and run with java -XX:-UseCompressedOops -jar tests-custom/target/jcstress.jar -t DCLFinal. Testing code below (mostly modified singleton testing classes):
/*
* SafeDCLFinal.java:
*/
package org.openjdk.jcstress.tests.singletons;
public class SafeDCLFinal {
#JCStressTest
#JCStressMeta(GradingSafe.class)
public static class Unsafe {
#Actor
public final void actor1(SafeDCLFinalFactory s) {
s.getInstance(SingletonUnsafe::new);
}
#Actor
public final void actor2(SafeDCLFinalFactory s, IntResult1 r) {
r.r1 = Singleton.map(s.getInstance(SingletonUnsafe::new));
}
}
#JCStressTest
#JCStressMeta(GradingSafe.class)
public static class Safe {
#Actor
public final void actor1(SafeDCLFinalFactory s) {
s.getInstance(SingletonSafe::new);
}
#Actor
public final void actor2(SafeDCLFinalFactory s, IntResult1 r) {
r.r1 = Singleton.map(s.getInstance(SingletonSafe::new));
}
}
#State
public static class SafeDCLFinalFactory {
private Singleton instance; // specifically non-volatile
public Singleton getInstance(Supplier<Singleton> s) {
if (instance == null) {
synchronized (this) {
if (instance == null) {
// instance = s.get();
instance = Publisher.publish(s.get(), true);
}
}
}
return instance;
}
}
}
/*
* UnsafeDCLFinal.java:
*/
package org.openjdk.jcstress.tests.singletons;
public class UnsafeDCLFinal {
#JCStressTest
#JCStressMeta(GradingUnsafe.class)
public static class Unsafe {
#Actor
public final void actor1(UnsafeDCLFinalFactory s) {
s.getInstance(SingletonUnsafe::new);
}
#Actor
public final void actor2(UnsafeDCLFinalFactory s, IntResult1 r) {
r.r1 = Singleton.map(s.getInstance(SingletonUnsafe::new));
}
}
#JCStressTest
#JCStressMeta(GradingUnsafe.class)
public static class Safe {
#Actor
public final void actor1(UnsafeDCLFinalFactory s) {
s.getInstance(SingletonSafe::new);
}
#Actor
public final void actor2(UnsafeDCLFinalFactory s, IntResult1 r) {
r.r1 = Singleton.map(s.getInstance(SingletonSafe::new));
}
}
#State
public static class UnsafeDCLFinalFactory {
private Singleton instance; // specifically non-volatile
public Singleton getInstance(Supplier<Singleton> s) {
if (instance == null) {
synchronized (this) {
if (instance == null) {
// instance = s.get();
instance = Publisher.publish(s.get(), false);
}
}
}
return instance;
}
}
}
/*
* Publisher.java:
*/
package org.openjdk.jcstress.tests.singletons;
public class Publisher {
public static <T> T publish(T val, boolean safe){
if(safe){
return new SafePublish<T>(val).get();
}
return new UnsafePublish<T>(val).get();
}
private static class UnsafePublish<T>{
T val;
public UnsafePublish(T val) {
this.val = val;
}
public T get(){
return val;
}
}
private static class SafePublish<T>{
final T val;
public SafePublish(T val) {
this.val = val;
}
public T get(){
return val;
}
}
}
Tested with java 8, but should work at least with java 6+. See docs
But I wonder if this would work:
// Double-check idiom for lazy initialization of instance fields without volatile
private FieldHolder fieldHolder = null;
private static class FieldHolder{
public final FieldType field;
FieldHolder(){
field = computeFieldValue();
}
}
FieldType getField() {
if (fieldHolder == null) { // First check (no locking)
synchronized(this) {
if (fieldHolder == null) // Second check (with locking)
fieldHolder = new FieldHolder();
}
}
return fieldHolder.field;
}
Or maybe even:
// Double-check idiom for lazy initialization of instance fields without volatile
private FieldType field = null;
private static class FieldHolder{
public final FieldType field;
FieldHolder(){
field = computeFieldValue();
}
}
FieldType getField() {
if (field == null) { // First check (no locking)
synchronized(this) {
if (field == null) // Second check (with locking)
field = new FieldHolder().field;
}
}
return field;
}
Or:
// Double-check idiom for lazy initialization of instance fields without volatile
private FieldType field = null;
FieldType getField() {
if (field == null) { // First check (no locking)
synchronized(this) {
if (field == null) // Second check (with locking)
field = new Object(){
public final FieldType field = computeFieldValue();
}.field;
}
}
return field;
}
I belive this would work based on this oracle doc:
The usage model for final fields is a simple one: Set the final fields for an object in that object's constructor; and do not write a reference to the object being constructed in a place where another thread can see it before the object's constructor is finished. If this is followed, then when the object is seen by another thread, that thread will always see the correctly constructed version of that object's final fields. It will also see versions of any object or array referenced by those final fields that are at least as up-to-date as the final fields are.
First things first: what you are trying to do is dangerous at best. I am getting a bit nervous when people try to cheat with finals. Java language provides you with volatile as the go-to tool to deal with inter-thread consistency. Use it.
Anyhow, the relevant approach is described in
"Safe Publication and Initialization in Java" as:
public class FinalWrapperFactory {
private FinalWrapper wrapper;
public Singleton get() {
FinalWrapper w = wrapper;
if (w == null) { // check 1
synchronized(this) {
w = wrapper;
if (w == null) { // check2
w = new FinalWrapper(new Singleton());
wrapper = w;
}
}
}
return w.instance;
}
private static class FinalWrapper {
public final Singleton instance;
public FinalWrapper(Singleton instance) {
this.instance = instance;
}
}
}
It layman's terms, it works like this. synchronized yields the proper synchronization when we observe wrapper as null -- in other words, the code would be obviously correct if we drop the first check altogether and extend synchronized to the entire method body. final in FinalWrapper guarantees iff we saw the non-null wrapper, it is fully constructed, and all Singleton fields are visible -- this recovers from the racy read of wrapper.
Note that it carries over the FinalWrapper in the field, not the value itself. If instance were to be published without the FinalWrapper, all bets would be off (in layman terms, that's premature publication). This is why your Publisher.publish is disfunctional: just putting the value through final field, reading it back, and publishing it unsafely is not safe -- it's very similar to just putting the naked instance write out.
Also, you have to be careful to make a "fallback" read under the lock, when you discover the null wrapper, and use its value. Doing the second (third) read of wrapper in return statement would also ruin the correctness, setting you up for a legitimate race.
EDIT: That entire thing, by the way, says that if the object you are publishing is covered with final-s internally, you may cut the middleman of FinalWrapper, and publish the instance itself.
EDIT 2: See also, LCK10-J. Use a correct form of the double-checked locking idiom, and some discussion in comments there.
In short
The version of the code without the volatile or the wrapper class is dependent on the memory model of the underlying operating system that the JVM is running on.
The version with the wrapper class is a known alternative known as the Initialization on Demand Holder design pattern and relies upon the ClassLoader contract that any given class is loaded at most once, upon first access, and in a thread-safe way.
The need for volatile
The way developers think of code execution most of the time is that the program is loaded into main memory and directly executed from there. The reality, however, is that there are a number of hardware caches between main memory and the processor cores. The problem arises because each thread might run on separate processors, each with their own independent copy of the variables in scope; while we like to logically think of field as a single location, the reality is more complicated.
To run through a simple (though perhaps verbose) example, consider a scenario with two threads and a single level of hardware caching, where each thread has their own copy of field in that cache. So already there are three versions of field: one in main memory, one in the first copy, and one in the second copy. I'll refer to these as fieldM, fieldA, and fieldB respectively.
Initial statefieldM = nullfieldA = nullfieldB = null
Thread A performs the first null-check, finds fieldA is null.
Thread A acquires the lock on this.
Thread B performs the first null-check, finds fieldB is null.
Thread B tries to acquire the lock on this but finds that it's held by thread A. Thread B sleeps.
Thread A performs the second null-check, finds fieldA is null.
Thread A assigns fieldA the value fieldType1 and releases the lock. Since field is not volatile this assignment is not propagated out.fieldM = nullfieldA = fieldType1fieldB = null
Thread B awakes and acquires the lock on this.
Thread B performs the second null-check, finds fieldB is null.
Thread B assigns fieldB the value fieldType2 and releases the lock.fieldM = nullfieldA = fieldType1fieldB = fieldType2
At some point, the writes to cache copy A are synched back to main memory.fieldM = fieldType1fieldA = fieldType1fieldB = fieldType2
At some later point, the writes to cache copy B are synched back to main memory overwriting the assignment made by copy A.fieldM = fieldType2fieldA = fieldType1fieldB = fieldType2
As one of the commenters on the question mentioned, using volatile ensures writes are visible. I don't know the mechanism used to ensure this -- it could be that changes are propagated out to each copy, it could be that the copies are never made in the first place and all accesses of field are against main memory.
One last note on this: I mentioned earlier that the results are system dependent. This is because different underlying systems may take less optimistic approaches to their memory model and treat all memory shared across threads as volatile or may perhaps apply a heuristic to determine whether a particular reference should be treated as volatile or not, though at the cost of performance of synching to main memory. This can make testing for these problems a nightmare; not only do you have to run against a enough large sample to try to trigger the race condition, you might just happen to be testing on a system which is conservative enough to never trigger the condition.
Initialization on Demand holder
The main thing I wanted to point out here is that this works because we're essentially sneaking a singleton into the mix. The ClassLoader contract means that while there can many instances of Class, there can be only a single instance of Class<A> available for any type A, which also happens to be loaded on first when first reference / lazily-initialized. In fact, you can think of any static field in a class's definition as really being fields in a singleton associated with that class where there happens to be increased member access privileges between that singleton and instances of the class.
Quoting The "Double-Checked Locking is Broken" Declaration mentioned by #Kicsi, the very last section is:
Double-Checked Locking Immutable Objects
If Helper is an immutable object, such that all of the fields of
Helper are final, then double-checked locking will work without having
to use volatile fields. The idea is that a reference to an immutable
object (such as a String or an Integer) should behave in much the same
way as an int or float; reading and writing references to immutable
objects are atomic.
(emphasis is mine)
Since FieldHolder is immutable, you indeed don't need the volatile keyword: other threads will always see a properly-initialized FieldHolder. As far as I understand it, the FieldType will thus always be initialized before it can be accessed from other threads through FieldHolder.
However, proper synchronization remains necessary if FieldType is not immutable. By consequent I'm not sure you would have much benefit from avoiding the volatile keyword.
If it is immutable though, then you don't need the FieldHolder at all, following the above quotation.
Using Enum or nested static class helper for lazy initialization otherwise just use static initialization if the initialization won't take much cost (space or time).
public enum EnumSingleton {
/**
* using enum indeed avoid reflection intruding but also limit the ability of the instance;
*/
INSTANCE;
SingletonTypeEnum getType() {
return SingletonTypeEnum.ENUM;
}
}
/**
* Singleton:
* The JLS guarantees that a class is only loaded when it's used for the first time
* (making the singleton initialization lazy)
*
* Thread-safe:
* class loading is thread-safe (making the getInstance() method thread-safe as well)
*
*/
private static class SingletonHelper {
private static final LazyInitializedSingleton INSTANCE = new LazyInitializedSingleton();
}
The "Double-Checked Locking is Broken" Declaration
With this change, the Double-Checked Locking idiom can be made to work by declaring the helper field to be volatile. This does not work under JDK4 and earlier.
class Foo {
private volatile Helper helper = null;
public Helper getHelper() {
if (helper == null) {
synchronized(this) {
if (helper == null)
helper = new Helper();
}
}
return helper;
}
}
No, this would not work.
final does not guarantee the visibility between threads that volatile does. The Oracle doc you quoted says that other threads will always see a correctly constructed version of an object's final fields. final guarantees that all final fields have been constructed and set by the time an objects constructor has finished running. So if object Foo contains a final field bar, bar is guaranteed to be constructed by the time Foo's constructor has finished.
The object referenced by a final field is still mutable though, and writes to that object may not be correctly visible across different threads.
So in your examples, other threads are not guaranteed to see the FieldHolder object that has been created and may create another, or if any modifications happen to the state of the FieldType object, it is not guaranteed that other threads will see these modifications. The final keyword is only guaranteeing that once the other threads do see the FieldType object, its constructor has been called.
So we were recently introduced to the Singleton Pattern, and I've found a lot of great uses for it since, but there's just one really small thing that bugs me about it. I've noticed that in every example I've seen for the singleton pattern,the standard getInstance() method looks like this:
private static Singleton instance = null;
public static Singleton getInstance() {
if ( instance == null )
instance = new Singleton();
return instance;
}
The thing I want to know is, is there a point in checking if the instance variable is null?
Would it work if you just assigned the instance variable to a new Singleton instance straight away and just return it like this:
private static Singleton instance = new Singleton();
public static Singleton getInstance() {
return instance;
}
Sorry if this seems like a waste of time of a question, but I just wanted to know if there was any reason as to why the first example is use everywhere.
Thanks for your time.
EDIT: Forgot to make the methods static.
Both are valid ways of creating a Singleton instance. The former is called lazy initialization and the latter is called eager initialization
Lazy initialization will help if the cost of creation of singleton instance is high. In this case the singleton instance is created only when its required.
On the other hand, eager initialization is by default Thread-safe
The 1-st example (lazy load) is not thread safe; to make it thread safe you can put it
private static volatile Singleton instance = null;
private static final Object syncObj = new syncObj();
public static Singleton getInstance() { // <- do not forget "static"
// Double checked locking pattern
if (instance != null)
return instance;
synchronize(syncObj) {
if (instance != null)
return instance;
instance = new Singleton();
}
return instance;
}
The second example (eager load) is thread safe:
// "final" looks good here
private static final Singleton instance = new Singleton();
public static Singleton getInstance() { // <- do not forget "static"
return instance;
}
In the first example you only create an instance of the class when you really need it (lazy initialization) the first time. So, you are using less memory until you need the class.
If you have a lot of Singleton, you are saving memory and time if you finally don't need them.
First example is a lazy loading of instance creation where first request to getInstance method create the Instance.
To avoid multiple object creation using lazy loading at first time method call (from multiple thread at same time), Object creation must be synchronized.
if (null == instance ) {
synchronized (Singleton.class){
if (null == instance ) {
instance = new Singleton();
}
}
}
This question already has answers here:
Why is volatile used in double checked locking
(8 answers)
Closed 5 years ago.
private volatile static Singleton uniqueInstance
In a singleton when using double lock method for synchronization why is the single instance declared as volatile ? Can I achieve the same functionality without declaring it as volatile ?
The volatile prevents memory writes from being re-ordered, making it impossible for other threads to read uninitialized fields of your singleton through the singleton's pointer.
Consider this situation: thread A discovers that uniqueInstance == null, locks, confirms that it's still null, and calls singleton's constructor. The constructor makes a write into member XYZ inside Singleton, and returns. Thread A now writes the reference to the newly created singleton into uniqueInstance, and gets ready to release its lock.
Just as thread A gets ready to release its lock, thread B comes along, and discovers that uniqueInstance is not null. Thread B accesses uniqueInstance.XYZ thinking that it has been initialized, but because the CPU has reordered writes, the data that thread A has written into XYZ has not been made visible to thread B. Therefore, thread B sees an incorrect value inside XYZ, which is wrong.
When you mark uniqueInstance volatile, a memory barrier is inserted. All writes initiated prior to that of uniqueInstance will be completed before the uniqueInstance is modified, preventing the reordering situation described above.
Without volatile the code doesn't work correctly with multiple threads.
From Wikipedia's Double-checked locking:
As of J2SE 5.0, this problem has been fixed. The volatile keyword now ensures that multiple threads handle the singleton instance correctly. This new idiom is described in The "Double-Checked Locking is Broken" Declaration:
// Works with acquire/release semantics for volatile
// Broken under Java 1.4 and earlier semantics for volatile
class Foo {
private volatile Helper helper = null;
public Helper getHelper() {
Helper result = helper;
if (result == null) {
synchronized(this) {
result = helper;
if (result == null) {
helper = result = new Helper();
}
}
}
return result;
}
// other functions and members...
}
In general you should avoid double-check locking if possible, as it is difficult to get right and if you get it wrong it can be difficult to find the error. Try this simpler approach instead:
If the helper object is static (one per class loader), an alternative is the initialization on demand holder idiom
// Correct lazy initialization in Java
#ThreadSafe
class Foo {
private static class HelperHolder {
public static Helper helper = new Helper();
}
public static Helper getHelper() {
return HelperHolder.helper;
}
}
To avoid using double locking, or volatile I use the follow
enum Singleton {
INSTANCE;
}
Creating the instance is simple, lazy loaded and thread safe.
Write to a volatile field will happen before any read operation.
Below is an example code for better understanding:
private static volatile ResourceService resourceInstance;
//lazy Initialiaztion
public static ResourceService getInstance () {
if (resourceInstance == null) { // first check
synchronized(ResourceService.class) {
if (resourceInstance == null) { // double check
// creating instance of ResourceService for only one time
resourceInstance = new ResourceService ();
}
}
}
return resourceInstance;
}
This link can serve you better
http://javarevisited.blogspot.com/2011/06/volatile-keyword-java-example-tutorial.html
You can use the follow code:
private static Singleton uniqueInstance;
public static synchronized Singleton getInstance(){
if(uniqueInstance == null){
uniqueInstance = new Singleton();
}
return uniqueInstance
}